Doc: AMR mineral revenue and school funding data dictionary
Status: Complete local analytical package, pending parent page acceptance
Owner: Rene
Last verified: 2026-11-14
Depends on: evidence-contract.json, sources/source-ledger.json, analysis.py and original financial statements

This is a compilation of named public accounts, not a state ranking or a measurement of classroom spending. Texas FY2025 is September 1, 2024 through August 31, 2025. New Mexico FY2025 is July 1, 2024 through June 30, 2025. Exact source dollars are preserved with Decimal arithmetic. A missing field is never replaced by zero. Observed zero land transfers in the original beneficiary exhibit remain zero.

| Dataset | Population and purpose | Important qualification |
| --- | --- | --- |
| fund-crosswalk-2025.csv | Selected complete accounting measures, with entity, amount, period, source page and qualification | Rows are not additive. Texas governmental PSF, PSF Corporation and ASF are different entities. NM SLO deposits and SIC distributions are different steps. |
| receipt-category-comparison-2025.csv | Every exact receipt amount shown in Figure2, with audit basis and entity | Category scopes differ and do not support a state royalty ranking. |
| fund-balance-bridges-2025.csv | All opening stocks, annual component flows and closing stocks needed to close both endowment statements | Texas reports net position; NM reports fund balance. Signed outflows remain negative. NM investment income, federal grazing and administration remain separate. |
| nm-beneficiaries-2025.csv | Every one of the 21 accounts in audited SIC Exhibit 1, with all five dollar columns and participation | Public schools account includes additional education distribution. Other20 includes educational institutions. Other capital change is a net measure. |
| texas-distribution-components-2025.csv | Total corporation transfer to ASF, included SLB subset, AMR derived remainder | Never add the included subset to the reported total. |
| nm-land-office-audited-pathways-2025.csv | Every revenue line needed to reproduce SLO governmental total revenue, and its distribution endpoint | Includes negative reapportionment. The SLO distribution line includes invested trust deposits. |
| nm-administrative-receipts-and-rentals-2025.csv | Complete selected source totals and rental-distribution totals used by the article | Administrative, not audited. Rental distributions include prior FY2024 reapportionment. Source royalty category includes land contracts. |
| texas-unaudited-receipt-detail-2025.csv | Exact selected GLO Schedule4 lines used to explain the receipt boundary | Explicitly unaudited and not an exhaustive receipt total. Negative signs retained. |
| texas-asf-distributions-2016-2025.csv | Every annual row in the source's ten-year ASF destination series | Statistical schedule expressly unaudited; source in USD thousands. Preserves predecessor reporting entities and FY2023 transition. |
| derived-shares-2025.csv | Explicit numerators, denominators and AMR calculated shares | Percent rounded half up to two decimals; no royalty-to-classroom or interstate efficiency ratio. |
| reconciliation-register.csv | Four reproducible paired-source comparisons and differences | Preserves NM $1 inter-source difference and TX $137 source-thousands rounding difference. |
| newer-announcement-separate.csv | FY2026 approximate agency revenue and nonroyalty lower bound | Rounded announcement, not audited revenue breakdown or school expenditure. Do not calculate exact residual mineral revenue. |
| source-cells.csv | All71 numerical observations with source provenance and original precision | A download-only source ledger, not an additive analytical table. |

`amount_usd`, `value` and `signed_usd` contain source or derived dollars as labeled. `period` is an ISO interval, not the retrieval date. `sequence` orders the exact bridge. `role` distinguishes a stock at the period endpoint from annual flows. `basis` states whether a source is audited, an administrative schedule, an unaudited agency schedule or a rounded announcement. `source_id` joins the original producer record in source-ledger.json; `source_page` gives the printed page, plus PDF page where verified. `source_vintage` preserves the source's own date; source hashes identify the retained original capture. The Texas GLO scanned transmittal is November1,2025, regardless of the December upload directory.

In the beneficiary table, `beginning_usd` and `ending_usd` are account stocks. `slo_transfer_usd` is the incoming trust allocation. `distribution_usd` preserves the source negative sign, while `distribution_magnitude_usd` supplies a positive magnitude for a clear percentage denominator. `other_capital_change_usd` preserves “Other Increases in Capital Accounts, Net,” not pure investment earnings. `source_rounding_residual_usd` is ending minus the sum of opening and all source flows; every row closes exactly in this extract. `reported_participation_percent` is the original five-decimal asset participation percentage, distinct from AMR's two-decimal `distribution_share_percent` and `asset_share_calculated_percent`.

The reconciliation register's `difference` is right value minus left value. A positive NM residual does not identify a cause. The derived share numerator and denominator always belong to the same account population, column and period. The other 20 NM accounts are never classified as all noneducation. Texas history retains original thousands and separately supplies converted dollars; the 2025 history figure is not silently changed to the more precise current audited amount.

Reproduce with `python3 analysis.py`, then `python3 figures.py`. reproduction-inputs.json lists exact offline input files. The source PDFs are kept locally for independent review, but are unnecessary for routine regeneration. Original sources retain their rights and credit. AMR's compilation, calculations and original figures are offered under the license in metadata.json; this does not relicense third-party reports or native font assets.
